Everyone Matters: A Multimodal Learning Analysis Framework Based on Individual Time Series

Multimodal learning analysis emphasizes using diverse data from various sources and forms for precise examination of learning patterns. Despite recent rapid advancements in this field, conventional learning analysis remains predominantly cross-sectional and group-focused, which is insufficient for understanding continuous and personalized learning processes, especially in multimodal situations. Moreover, multimodal learning analysis poses a significant technological challenge, with research indicating growing difficulties for educators. Thus, we introduce an innovative framework utilizing individual learners’ temporal sequences for multimodal learning analysis, offering multifaceted evidence for understanding individual learning journeys. We elucidate the analysis process and extend the framework to pre-service teachers (Case 1) and in-service educators (Case 2), promoting their involvement in advancing multimodal learning analysis. Case findings underscore the framework’s feasibility and comprehensibility, suggesting its potential to drive personalized learning analysis in future practices and research endeavors.